NERDSS Development

Structure-Resolved Reaction-Diffusion Simulation Software by Johnson Lab, JHU

Installation

---

Build Serial NERDSS

Step 1 — Install a C++ Compiler

macOS

Install Xcode via App Store. Alternatively, use commandline:

xcode-select --install

Ubuntu / Linux

sudo apt update
sudo apt install build-essential

---

Step 2 — Install GNU Scientific Library (GSL ≥ 1.0)

macOS (Homebrew)

brew install gsl

Ubuntu / Linux

sudo apt install libgsl-dev

---

Step 3 — Compile Serial Version

From the NERDSS root directory:

make serial

Executable will appear in:

./bin

---

Build Parallel NERDSS (MPI)

Step 1 — Switch to the MPI Branch

git checkout mpi

---

Step 2 — Install MPI (OpenMPI)

macOS (Homebrew)

brew install open-mpi

Ubuntu / Linux

sudo apt install openmpi-bin libopenmpi-dev

---

Step 3 — Compile MPI Version

From the NERDSS root directory:

make mpi

Executable will appear in:

./bin

---

Quick Verification

After building, check:

ls ./bin

You should see the compiled executable(s).

---

Summary

Build Type	Command
Serial	make serial
MPI	git checkout mpi → make mpi

---

Run Simulations

Run a quick trial with Google Colab

Click the following link to make a copy of the iPython notebook in your Google Colab and following the instructions on the Notebook (Note currently the link is under NERDSS development repo. The link will need to be updated once synced to released repo)

Compile and run NERDSS on your local machine

1. Example input files are located in the subdirectories within the sample_inputs folder.

2. To start the serial simulation, use the command ./nerdss -f parms.inp.

3. To start the parallel simulation, use the command mpirun -np $nprocs ./nerdss_mpi -f parms.inp.

4. To debug the parallel code, use the command mpirun -np 2 xterm -e gdb --ex 'b error' --ex r --args ./nerdss_mpi -f parms.inp -s 123.

Analyzing Results

1. Use the ioNERDSS PyPi library for visualizing simulation results.
2. Install ioNERDSS with pip install ioNERDSS.
3. Refer to the ionerdss repository for more details.

Style Guides for Developers

1. We use the Google C++ Style Guide and the C++ Core Guidelines and prefer the C++ Standard Library and modern C++ features.
2. Provide comments and documentation to explain complex or non-obvious code sections. Write unit tests to ensure code quality.